논문 영어

1. To understand the impact of explanations on decision making when using a KBS, several dimensions must be considered.  
2. While other effects beyond the scope of this study may occur, the effects of particular interest in the current study are the bolded items in Figure 1.
3. Dhaliwal and Benbasat (1996) provide the original theoretical basis for integrating feedforward/feedback modes with different explanation types. 
4. Arnold et al. (2004a) expand on these six forms of explanation to include the other two combinations:  feedforward/definition and feedback/ definition.
5. The model presented in Figure 1 aligns closely with the theory provided by Gregor and Benbasat (1999), particularly with the following propositions:
6. The key is to examine these propositions within the context of a decision-making scenario where the behavior of the decision maker can be observed while working in a cooperative problem solving mode with the system.
7. This expected effect on novices is consistent with the theory of technology dominance (Arnold and Sutton 1998), which posits that novices who have a limited capability to make a judgment within a given domain will be highly likely to rely on the recommendations of a KBS with or without explanations.
8. The theory implies that the KBS, rather than the user, dominates the decision-making process.
9. The theory posits that a KBS will influence an expert’s decision, but an expert may not adhere to the recommendations of the aid.
10. Evidence of such novice and expert behavior has been found in recent studies (Arnold et al. 2004b; Masselli et al. 2002; Noga and Arnold 2002).
11. This leads to the first hypothesis.
12. KBS users will be more likely to adhere to the recommendation of a KBS when explanations are provided.
13. Several factors that appear to impact explanation preferences and use are also likely to affect decision making outcomes.
14. Subsequently, Mao and Benbasat (2000) focused on expert versus novice differences in using and processing explanations.
15. They anticipated that the substantial differences in cognitive structures and processes used by experts versus novices would carry over to use of a KBS.
16. Their results confirmed their expectation that novice users focused heavily on explanations that aided in understanding the reasoning process.
17. A major theoretical contribution of their work was that domain expertise could influence explanation use.
18. Multiple dimensions can explain the differences between novice and expert users. 
19. A feedforward explanation is described as declarative because it provides an explanation about how inputs to a KBS are used in terms of relevant information cues and their relationships.
20. Conversely, a feedback explanation is tailored to describe processing in terms of